QC should be applied risk-based to each data handling stage, with monitoring and data management as the main QC activities and QC extendable to facilities such as central image readers. Monitoring protects participants and result reliability through communication with sites, checking qualifications and resources, training, and review of documents and data by source data review, source data verification, analytics and visits; it should be done by people not involved in clinical conduct at the monitored site, respect data protection, consider decentralised elements, and be set out in the monitoring plan. The sponsor should decide the extent and nature of monitoring from identified risks and trial factors (objective, design, complexity, blinding, size, product, safety knowledge, endpoints). Site monitoring, with frequency based on risk and adjusted with experience, may be on-site or remote, including secure read-only remote access to source and records; centralised monitoring by qualified staff (medical monitors, data scientists, biostatisticians) evaluates accumulating data promptly, can complement or replace some site monitoring, can reveal systemic or site-specific issues and can target site visits. The monitoring plan should be tailored to safety, data quality and reliability risks, focus on critical aspects and procedures for safety and endpoints, describe strategy, all parties' activities, methods, tools and rationale, consider site capability and burden, reference sponsor procedures and cover important data and processes outside sites; monitors follow the plan and procedures.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.